Improving the Diversity of Unsupervised Paraphrasing with Embedding Outputs

We present a novel technique for zero-shot paraphrase generation. The key contribution is an end-to-end multilingual paraphrasing model that is trained using translated parallel corpora to generate paraphrases into meaning spaces'' -- replacing the final softmax layer with word embeddings. This architectural modification, plus a training procedure that incorporates an autoencoding objective, enables effective parameter sharing across languages for more fluent monolingual rewriting, and facilitates fluency and diversity in the generated outputs. Our continuous-output paraphrase generation models outperform zero-shot paraphrasing baselines when evaluated on two languages using a battery of computational metrics as well as in human assessment.

Informed Sampling for Diversity in Concept-to-Text NLG

Deep-learning models for language generation tasks tend to produce repetitive output. Various methods have been proposed to encourage lexical diversity during decoding, but this often comes at a cost to the perceived fluency and adequacy of the output. In this work, we propose to ameliorate this cost by using an Imitation Learning approach to explore the level of diversity that a language generation model can reliably produce. Specifically, we augment the decoding process with a meta-classifier trained to distinguish which words at any given timestep will lead to high-quality output. We focus our experiments on concept-to-text generation where models are sensitive to the inclusion of irrelevant words due to the strict relation between input and output. Our analysis shows that previous methods for diversity underperform in this setting, while human evaluation suggests that our proposed method achieves a high level of diversity with minimal effect on the output's fluency and adequacy.

Grouping Words with Semantic Diversity

Deep Learning-based NLP systems can be sensitive to unseen tokens and hard to learn with high-dimensional inputs, which critically hinder learning generalization. We introduce an approach by grouping input words based on their semantic diversity to simplify input language representation with low ambiguity. Since the semantically diverse words reside in different contexts, we are able to substitute words with their groups and still distinguish word meanings relying on their contexts. We design several algorithms that compute diverse groupings based on random sampling, geometric distances, and entropy maximization, and we prove formal guarantees for the entropy-based algorithms. Experimental results show that our methods generalize NLP models and demonstrate enhanced accuracy on POS tagging and LM tasks and significant improvements on medium-scale machine translation tasks, up to +6.5 BLEU points. Our source code is available at https://github.com/abdulrafae/dg.

EaSe: A Diagnostic Tool for VQA based on Answer Diversity

We propose EASE, a simple diagnostic tool for Visual Question Answering (VQA) which quantifies the difficulty of an image, question sample. EASE is based on the pattern of answers provided by multiple annotators to a given question. In particular, it considers two aspects of the answers: (i) their Entropy; (ii) their Semantic content. First, we prove the validity of our diagnostic to identify samples that are easy/hard for state-of-art VQA models. Second, we show that EASE can be successfully used to select the most-informative samples for training/fine-tuning. Crucially, only information that is readily available in any VQA dataset is used to compute its scores.

No NLP Task Should be an Island: Multi-disciplinarity for Diversity in News Recommender Systems

Natural Language Processing (NLP) is defined by specific, separate tasks, with each their own literature, benchmark datasets, and definitions. In this position paper, we argue that for a complex problem such as the threat to democracy by non-diverse news recommender systems, it is important to take into account a higher-order, normative goal and its implications. Experts in ethics, political science and media studies have suggested that news recommendation systems could be used to support a deliberative democracy. We reflect on the role of NLP in recommendation systems with this specific goal in mind and show that this theory of democracy helps to identify which NLP tasks and techniques can support this goal, and what work still needs to be done. This leads to recommendations for NLP researchers working on this specific problem as well as researchers working on other complex multidisciplinary problems.

Multiple Input Multiple Output Comm. Systems “MIMO”

إنّ تزايد الطلب على معدلات النقل في الاتصالات دفع الباحثين لإيجاد طرق جديدة لتلبيّة هذه المتطلبات، ومع محدودية المصادر الترددية (عرض الحزمة) والزمنية تمّ التوّجه لإيجاد مصادر جديدة للنظام تساعد على زيادة معدل النقل وتحسين جودة النظام في الوقت نفسه بغية توفير عرض الحزمة والاستطاعة المستهلكة في الأنظمة التقليدية الحالية التي تحوي هوائي إرسال واحد وهوائي استقبال واحد. ظهرت أنظمة الاتصالات متعددة المداخل والمخارج MIMO لتقدّم الكثير من التحسينات والميّزات الجديدة لشبكات الاتصالات اللاسلكيّة وأدخلت أنظمة الاتصالات بعصر جديد يستخدم النظام فيه عدة هوائيات في الإرسال والاستقبال للاستفادة من خصائص القناة متعددة المسارات Multipath Channel عبر تقديم ربحين: ربح التنويع Diversity Gain وربح التنضيد Multiplexing Gain حيث يعمل ربح التنويع على زيادة تحسين جودة وأداء النظام، بينما يعمل ربح التنضيد على زيادة معدل نقل النظام، كما أدخلت مؤخرا ميزات أخرى على هذا النظام أهمها تقنية تشكيل الحزمة Beamforming التي تعمل على تركيز الحزمة بالاتجاه المطلوب لتقليل التداخل بين الأجهزة وزيادة جودة الإشارة. وظفت هذه الميزات في الكثير من أنظمة الاتصالات ومنها أنظمة Wi-Fi والأنظمة الخلوية LTE و5G التي أدخلت مفاهيم جديدة على أنظمة MIMO أهمها: تقنية Massive MIMO الذي ساعد على تحقيق معدلات وكفاءات فائقة وتقنية Virtual MIMO التي تعمل على تشكيل مصفوفات إرسال/استقبال من عدة أجهزة متفرقة. سوف نقوم في هذا البحث بشرح أساسيات القناة متعددة المسارات وأنظمة MIMO، كيفية نمذجتها رياضياً، ميزات وأرباح هذه المنظومة، بنية المنظومة، أهم تطبيقات هذه المنظومة وخصائصها الجديدة وأخيراً تحدياتها وطرق تطويرها.

Molecular Characterization of Jabali Goats Using SSR

The wide geographical spread and the exponential growth of the numbers of goats around the world clearly demonstrate the ability of these ruminants to adapt to harsh climates and grazing land. The aim of this study was to assess the genetic diversity of 42 samples of Syrian goats from many domestication stations including Jabali, Shami and hybrid (hybridization between both Shami and Jabali). The study was done by DNA extraction of these samples, application of SSR technology using 7 microstellite markers. The alleles number of markers were 29 alleles, at a rate of 4.1 allele for each genetic locus. The number of alleles of each locus ranged from 3 alleles in genetically markers (BMS1714, INRAD07, SRCRSP09) to 8 alleles at the genetic marker SRCRSP01. The rate of allele-frequency across all studied genetic sites ranged from 0.071429 with the molecular size of 210 bp for the McM527 genetic site to 0.97619 with a molecular size of 55 bp for the SRCRSP09 genetic site.

The role of economic reform in the structural composition of foreign trade in Syria during the period 0202 - 0222

The research aims to identify the role that foreign trade plays in the Syrian economy, They are an important part of the national economy, and which is secured need the economy to build production capacity of capital goods, and through the discharge of the flooding of the need for domestic demand of the product, and therefore the external sector developments reflect the productive structure of the national economy, employment and price and income level. . and also the effects of economic policies. The researcher has addressed the economic reforms that directly affect the foreign trade, and then analyzed the structure of the foreign trade to discover the real economic potentials and the necessary requirements for the promotion.

Study to determine the Asteraceae plants in the east of Golan

The search includes determining genus and species according to the flower's type and outside forms. So , 24 land medical species belong to 20 genus, all from Asteraceae, that is according to some references, registries and Atlas especially that one which studied turning this region into flowering one.

Molecular Characterization and Genetic Relatedness Between some Accessions of Origanum syriacum in Syria Using RAPD Technique

Randomly amplified polymorphic DNA (RAPD) technique was used to study the genetic relationship among 10 wild accessions of Origanum syriacum grown in Syria. The results of this study may have a great impact on the national biodiversity program which lacks the employment of molecular techniques, and thus we recommend making use of these techniques in genotype identification in general, and studying the relationship among them, and in particular in studying important wild species of medical importance which have not been described yet on the molecular level.